diff options
Diffstat (limited to 'nuttx/arch/arm/src/tiva/lm3s_ethernet.c')
| -rw-r--r-- | nuttx/arch/arm/src/tiva/lm3s_ethernet.c | 1478 |
1 files changed, 1478 insertions, 0 deletions
diff --git a/nuttx/arch/arm/src/tiva/lm3s_ethernet.c b/nuttx/arch/arm/src/tiva/lm3s_ethernet.c new file mode 100644 index 000000000..22cb53d8a --- /dev/null +++ b/nuttx/arch/arm/src/tiva/lm3s_ethernet.c @@ -0,0 +1,1478 @@ +/**************************************************************************** + * arch/arm/src/tiva/lm3s_ethernet.c + * + * Copyright (C) 2009-2010, 2014 Gregory Nutt. All rights reserved. + * Author: Gregory Nutt <gnutt@nuttx.org> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * 3. Neither the name NuttX nor the names of its contributors may be + * used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS + * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED + * AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN + * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + * + ****************************************************************************/ + +/**************************************************************************** + * Included Files + ****************************************************************************/ + +#include <nuttx/config.h> +#if defined(CONFIG_NET) && defined(CONFIG_TIVA_ETHERNET) + +#include <stdint.h> +#include <stdbool.h> +#include <time.h> +#include <string.h> +#include <debug.h> +#include <errno.h> + +#include <arpa/inet.h> + +#include <nuttx/arch.h> +#include <nuttx/wdog.h> +#include <nuttx/irq.h> +#include <arch/board/board.h> +#include <nuttx/net/arp.h> +#include <nuttx/net/netdev.h> + +#include "chip.h" +#include "up_arch.h" + +#include "tiva_gpio.h" +#include "tiva_ethernet.h" +#include "chip/tiva_pinmap.h" + +/**************************************************************************** + * Pre-processor Definitions + ****************************************************************************/ +/* Configuration ************************************************************/ + +#ifdef CONFIG_NET_MULTIBUFFER +# error CONFIG_NET_MULTIBUFFER should not be selected +#endif + +/* Half duplex can be forced if CONFIG_TIVA_ETHHDUPLEX is defined. */ + +#ifdef CONFIG_TIVA_ETHHDUPLEX +# define TIVA_DUPLEX_SETBITS 0 +# define TIVA_DUPLEX_CLRBITS MAC_TCTL_DUPLEX +#else +# define TIVA_DUPLEX_SETBITS MAC_TCTL_DUPLEX +# define TIVA_DUPLEX_CLRBITS 0 +#endif + +/* Auto CRC generation can be suppressed if CONFIG_TIVA_ETHNOAUTOCRC is definde */ + +#ifdef CONFIG_TIVA_ETHNOAUTOCRC +# define TIVA_CRC_SETBITS 0 +# define TIVA_CRC_CLRBITS MAC_TCTL_CRC +#else +# define TIVA_CRC_SETBITS MAC_TCTL_CRC +# define TIVA_CRC_CLRBITS 0 +#endif + +/* Tx padding can be suppressed if CONFIG_TIVA_ETHNOPAD is defined */ + +#ifdef CONFIG_TIVA_ETHNOPAD +# define TIVA_PADEN_SETBITS 0 +# define TIVA_PADEN_CLRBITS MAC_TCTL_PADEN +#else +# define TIVA_PADEN_SETBITS MAC_TCTL_PADEN +# define TIVA_PADEN_CLRBITS 0 +#endif + +#define TIVA_TCTCL_SETBITS (TIVA_DUPLEX_SETBITS|TIVA_CRC_SETBITS|TIVA_PADEN_SETBITS) +#define TIVA_TCTCL_CLRBITS (TIVA_DUPLEX_CLRBITS|TIVA_CRC_CLRBITS|TIVA_PADEN_CLRBITS) + +/* Multicast frames can be enabled by defining CONFIG_TIVA_MULTICAST */ + +#ifdef CONFIG_TIVA_MULTICAST +# define TIVA_AMUL_SETBITS MAC_RCTL_AMUL +# define TIVA_AMUL_CLRBITS 0 +#else +# define TIVA_AMUL_SETBITS 0 +# define TIVA_AMUL_CLRBITS MAC_RCTL_AMUL +#endif + +/* Promiscuous mode can be enabled by defining CONFIG_TIVA_PROMISCUOUS */ + +#ifdef CONFIG_TIVA_PROMISCUOUS +# define TIVA_PRMS_SETBITS MAC_RCTL_PRMS +# define TIVA_PRMS_CLRBITS 0 +#else +# define TIVA_PRMS_SETBITS 0 +# define TIVA_PRMS_CLRBITS MAC_RCTL_PRMS +#endif + +/* Bad CRC rejection can be enabled by define CONFIG_TIVA_BADCRC */ + +#ifdef CONFIG_TIVA_BADCRC +# define TIVA_BADCRC_SETBITS MAC_RCTL_BADCRC +# define TIVA_BADCRC_CLRBITS 0 +#else +# define TIVA_BADCRC_SETBITS 0 +# define TIVA_BADCRC_CLRBITS MAC_RCTL_BADCRC +#endif + +#define TIVA_RCTCL_SETBITS (TIVA_AMUL_SETBITS|TIVA_PRMS_SETBITS|TIVA_BADCRC_SETBITS) +#define TIVA_RCTCL_CLRBITS (TIVA_AMUL_CLRBITS|TIVA_PRMS_CLRBITS|TIVA_BADCRC_CLRBITS) + +/* CONFIG_TIVA_DUMPPACKET will dump the contents of each packet to the console. */ + +#ifdef CONFIG_TIVA_DUMPPACKET +# define tiva_dumppacket(m,a,n) lib_dumpbuffer(m,a,n) +#else +# define tiva_dumppacket(m,a,n) +#endif + +/* TX poll deley = 1 seconds. CLK_TCK is the number of clock ticks per second */ + +#define TIVA_WDDELAY (1*CLK_TCK) +#define TIVA_POLLHSEC (1*2) + +/* TX timeout = 1 minute */ + +#define TIVA_TXTIMEOUT (60*CLK_TCK) + +/* This is a helper pointer for accessing the contents of the Ethernet header */ + +#define ETHBUF ((struct eth_hdr_s *)priv->ld_dev.d_buf) + +#define TIVA_MAX_MDCCLK 2500000 + +/**************************************************************************** + * Private Types + ****************************************************************************/ + +/* EMAC statistics (debug only) */ + +#if defined(CONFIG_DEBUG) && defined(CONFIG_DEBUG_NET) +struct tiva_statistics_s +{ + uint32_t rx_int; /* Number of Rx interrupts received */ + uint32_t rx_packets; /* Number of packets received (sum of the following): */ + uint32_t rx_ip; /* Number of Rx IP packets received */ + uint32_t rx_arp; /* Number of Rx ARP packets received */ + uint32_t rx_dropped; /* Number of dropped, unsupported Rx packets */ + uint32_t rx_pktsize; /* Number of dropped, too small or too big */ + uint32_t rx_errors; /* Number of Rx errors (reception error) */ + uint32_t rx_ovrerrors; /* Number of Rx FIFO overrun errors */ + uint32_t tx_int; /* Number of Tx interrupts received */ + uint32_t tx_packets; /* Number of Tx packets queued */ + uint32_t tx_errors; /* Number of Tx errors (transmission error)*/ + uint32_t tx_timeouts; /* Number of Tx timeout errors */ +}; +# define EMAC_STAT(priv,name) priv->ld_stat.name++ +#else +# define EMAC_STAT(priv,name) +#endif + +/* The tiva_driver_s encapsulates all state information for a single hardware + * interface + */ + +struct tiva_driver_s +{ + /* The following fields would only be necessary on chips that support + * multiple Ethernet controllers. + */ + +#if TIVA_NETHCONTROLLERS > 1 + uint32_t ld_base; /* Ethernet controller base address */ + int ld_irq; /* Ethernet controller IRQ */ +#endif + + bool ld_bifup; /* true:ifup false:ifdown */ + WDOG_ID ld_txpoll; /* TX poll timer */ + WDOG_ID ld_txtimeout; /* TX timeout timer */ + +#if defined(CONFIG_DEBUG) && defined(CONFIG_DEBUG_NET) + struct tiva_statistics_s ld_stat; +#endif + + /* This holds the information visible to uIP/NuttX */ + + struct net_driver_s ld_dev; /* Interface understood by uIP */ +}; + +/**************************************************************************** + * Private Data + ****************************************************************************/ + +static struct tiva_driver_s g_lm3sdev[TIVA_NETHCONTROLLERS]; + +/**************************************************************************** + * Private Function Prototypes + ****************************************************************************/ + +/* Miscellaneous low level helpers */ + +#if TIVA_NETHCONTROLLERS > 1 +static uint32_t tiva_ethin(struct tiva_driver_s *priv, int offset); +static void tiva_ethout(struct tiva_driver_s *priv, int offset, uint32_t value); +#else +static inline uint32_t tiva_ethin(struct tiva_driver_s *priv, int offset); +static inline void tiva_ethout(struct tiva_driver_s *priv, int offset, uint32_t value); +#endif +static void tiva_ethreset(struct tiva_driver_s *priv); +#if 0 /* Not used */ +static void tiva_phywrite(struct tiva_driver_s *priv, int regaddr, uint16_t value); +#endif +static uint16_t tiva_phyread(struct tiva_driver_s *priv, int regaddr); + +/* Common TX logic */ + +static int tiva_transmit(struct tiva_driver_s *priv); +static int tiva_txpoll(struct net_driver_s *dev); + +/* Interrupt handling */ + +static void tiva_receive(struct tiva_driver_s *priv); +static void tiva_txdone(struct tiva_driver_s *priv); +static int tiva_interrupt(int irq, FAR void *context); + +/* Watchdog timer expirations */ + +static void tiva_polltimer(int argc, uint32_t arg, ...); +static void tiva_txtimeout(int argc, uint32_t arg, ...); + +/* NuttX callback functions */ + +static int tiva_ifup(struct net_driver_s *dev); +static int tiva_ifdown(struct net_driver_s *dev); +static int tiva_txavail(struct net_driver_s *dev); +#ifdef CONFIG_NET_IGMP +static int tiva_addmac(struct net_driver_s *dev, FAR const uint8_t *mac); +static int tiva_rmmac(struct net_driver_s *dev, FAR const uint8_t *mac); +#endif + +/**************************************************************************** + * Private Functions + ****************************************************************************/ + +/**************************************************************************** + * Function: tiva_ethin + * + * Description: + * Read a register from the Ethernet module + * + * Parameters: + * priv - Reference to the driver state structure + * offset - Byte offset of the register from the ethernet base address + * + * Returned Value: + * Register value + * + ****************************************************************************/ + +#if TIVA_NETHCONTROLLERS > 1 +static uint32_t tiva_ethin(struct tiva_driver_s *priv, int offset) +{ + return getreg32(priv->ld_base + offset); +} +#else +static inline uint32_t tiva_ethin(struct tiva_driver_s *priv, int offset) +{ + return getreg32(TIVA_ETHCON_BASE + offset); +} +#endif + +/**************************************************************************** + * Function: tiva_ethout + * + * Description: + * Write a register to the Ethernet module + * + * Parameters: + * priv - Reference to the driver state structure + * offset - Byte offset of the register from the ethernet base address + * value - The value to write the Ethernet register + * + * Returned Value: + * None + * + ****************************************************************************/ + +#if TIVA_NETHCONTROLLERS > 1 +static void tiva_ethout(struct tiva_driver_s *priv, int offset, uint32_t value) +{ + putreg32(value, priv->ld_base + offset); +} +#else +static inline void tiva_ethout(struct tiva_driver_s *priv, int offset, uint32_t value) +{ + putreg32(value, TIVA_ETHCON_BASE + offset); +} +#endif + +/**************************************************************************** + * Function: tiva_ethreset + * + * Description: + * Configure and reset the Ethernet module, leaving it in a disabled state. + * + * Parameters: + * priv - Reference to the driver state structure + * + * Returned Value: + * OK on success; a negated errno on failure + * + * Assumptions: + * + ****************************************************************************/ + +static void tiva_ethreset(struct tiva_driver_s *priv) +{ + irqstate_t flags; + uint32_t regval; + +#if TIVA_NETHCONTROLLERS > 1 +# error "If multiple interfaces are supported, this function would have to be redesigned" +#endif + + /* Make sure that clocking is enabled for the Ethernet (and PHY) peripherals */ + + flags = irqsave(); + regval = getreg32(TIVA_SYSCON_RCGC2); + regval |= (SYSCON_RCGC2_EMAC0|SYSCON_RCGC2_EPHY0); + putreg32(regval, TIVA_SYSCON_RCGC2); + nllvdbg("RCGC2: %08x\n", regval); + + /* Put the Ethernet controller into the reset state */ + + regval = getreg32(TIVA_SYSCON_SRCR2); + regval |= (SYSCON_SRCR2_EMAC0|SYSCON_SRCR2_EPHY0); + putreg32(regval, TIVA_SYSCON_SRCR2); + + /* Wait just a bit. This is a much longer delay than necessary */ + + up_mdelay(2); + + /* Then take the Ethernet controller out of the reset state */ + + regval &= ~(SYSCON_SRCR2_EMAC0|SYSCON_SRCR2_EPHY0); + putreg32(regval, TIVA_SYSCON_SRCR2); + nllvdbg("SRCR2: %08x\n", regval); + + /* Wait just a bit, again. If we touch the ethernet too soon, we may busfault. */ + + up_mdelay(2); + + /* Enable Port F for Ethernet LEDs: LED0=Bit 3; LED1=Bit 2 */ + +#ifdef CONFIG_TIVA_ETHLEDS + /* Configure the pins for the peripheral function */ + + tiva_configgpio(GPIO_ETHPHY_LED0 | GPIO_STRENGTH_2MA | GPIO_PADTYPE_STD); + tiva_configgpio(GPIO_ETHPHY_LED1 | GPIO_STRENGTH_2MA | GPIO_PADTYPE_STD); +#endif + + /* Disable all Ethernet controller interrupts */ + + regval = tiva_ethin(priv, TIVA_MAC_IM_OFFSET); + regval &= ~MAC_IM_ALLINTS; + tiva_ethout(priv, TIVA_MAC_IM_OFFSET, regval); + + /* Clear any pending interrupts (shouldn't be any) */ + + regval = tiva_ethin(priv, TIVA_MAC_RIS_OFFSET); + tiva_ethout(priv, TIVA_MAC_IACK_OFFSET, regval); + irqrestore(flags); +} + +/**************************************************************************** + * Function: tiva_phywrite + * + * Description: + * Write a 16-bit word to a PHY register + * + * Parameters: + * priv - Reference to the driver state structure + * regaddr - Address of the PHY register to write + * value - The value to write to the register + * + * Returned Value: + * None + * + ****************************************************************************/ + +#if 0 /* Not used */ +static void tiva_phywrite(struct tiva_driver_s *priv, int regaddr, uint16_t value) +{ + /* Wait for any MII transactions in progress to complete */ + + while ((tiva_ethin(priv, TIVA_MAC_MCTL_OFFSET) & MAC_MCTL_START) != 0); + + /* Set up the data to be written */ + + DEBUGASSERT(value < MAC_MTXD_MASK); + tiva_ethout(priv, TIVA_MAC_MTXD_OFFSET, value); + + /* Set up the PHY register address and start the write operation */ + + regaddr <<= MAC_MCTL_REGADR_SHIFT; + DEBUGASSERT((regaddr & MAC_MTXD_MASK) == regaddr); + tiva_ethout(priv, TIVA_MAC_MCTL_OFFSET, regaddr | MAC_MCTL_WRITE | MAC_MCTL_START); + + /* Wait for the write transaction to complete */ + + while ((tiva_ethin(priv, TIVA_MAC_MCTL_OFFSET) & MAC_MCTL_START) != 0); +} +#endif + +/**************************************************************************** + * Function: tiva_phyread + * + * Description: + * Write a 16-bit word to a PHY register + * + * Parameters: + * priv - Reference to the driver state structure + * regaddr - Address of the PHY register to write + * value - The value to write to the register + * + * Returned Value: + * None + * + ****************************************************************************/ + +static uint16_t tiva_phyread(struct tiva_driver_s *priv, int regaddr) +{ + /* Wait for any MII transactions in progress to complete */ + + while ((tiva_ethin(priv, TIVA_MAC_MCTL_OFFSET) & MAC_MCTL_START) != 0); + + /* Set up the PHY register address and start the read operation */ + + regaddr <<= MAC_MCTL_REGADR_SHIFT; + DEBUGASSERT((regaddr & MAC_MTXD_MASK) == regaddr); + tiva_ethout(priv, TIVA_MAC_MCTL_OFFSET, regaddr | MAC_MCTL_START); + + /* Wait for the write transaction to complete */ + + while ((tiva_ethin(priv, TIVA_MAC_MCTL_OFFSET) & MAC_MCTL_START) != 0); + + /* Read and return the PHY data */ + + return (uint16_t)(tiva_ethin(priv, TIVA_MAC_MRXD_OFFSET) & MAC_MTRD_MASK); +} + +/**************************************************************************** + * Function: tiva_transmit + * + * Description: + * Start hardware transmission. Called either from the txdone interrupt + * handling or from watchdog based polling. + * + * Parameters: + * priv - Reference to the driver state structure + * + * Returned Value: + * OK on success; a negated errno on failure + * + ****************************************************************************/ + +static int tiva_transmit(struct tiva_driver_s *priv) +{ + irqstate_t flags; + uint32_t regval; + uint8_t *dbuf; + int pktlen; + int bytesleft; + int ret = -EBUSY; + + /* Verify that the hardware is ready to send another packet */ + + flags = irqsave(); + if ((tiva_ethin(priv, TIVA_MAC_TR_OFFSET) & MAC_TR_NEWTX) == 0) + { + /* Increment statistics */ + + EMAC_STAT(priv, tx_packets); + tiva_dumppacket("Transmit packet", priv->ld_dev.d_buf, priv->ld_dev.d_len); + + /* Transfer the packet into the Tx FIFO. The LS 16-bits of the first + * 32-bit word written to the Tx FIFO contains the Ethernet payload + * data length. That is the full length of the message (d_len) minus + * the size of the Ethernet header (14). + */ + + pktlen = priv->ld_dev.d_len; + nllvdbg("Sending packet, pktlen: %d\n", pktlen); + DEBUGASSERT(pktlen > ETH_HDRLEN); + + dbuf = priv->ld_dev.d_buf; + regval = (uint32_t)(pktlen - 14); + regval |= ((uint32_t)(*dbuf++) << 16); + regval |= ((uint32_t)(*dbuf++) << 24); + tiva_ethout(priv, TIVA_MAC_DATA_OFFSET, regval); + + /* Write all of the whole, 32-bit values in the middle of the packet */ + + for (bytesleft = pktlen - 2; bytesleft > 3; bytesleft -= 4, dbuf += 4) + { + /* Transfer a whole word from the user buffer. Note, the user + * buffer may be un-aligned. + */ + + tiva_ethout(priv, TIVA_MAC_DATA_OFFSET, *(uint32_t*)dbuf); + } + + /* Write the last, partial word in the FIFO */ + + if (bytesleft > 0) + { + /* Write the last word */ + + regval = 0; + switch (bytesleft) + { + case 0: + default: + break; + + case 3: + regval |= ((uint32_t)dbuf[2] << 16); + case 2: + regval |= ((uint32_t)dbuf[1] << 8); + case 1: + regval |= (uint32_t)dbuf[0]; + break; + } + + tiva_ethout(priv, TIVA_MAC_DATA_OFFSET, regval); + } + + /* Activate the transmitter */ + + tiva_ethout(priv, TIVA_MAC_TR_OFFSET, MAC_TR_NEWTX); + + /* Setup the TX timeout watchdog (perhaps restarting the timer) */ + + (void)wd_start(priv->ld_txtimeout, TIVA_TXTIMEOUT, tiva_txtimeout, 1, (uint32_t)priv); + ret = OK; + } + + irqrestore(flags); + return ret; +} + +/**************************************************************************** + * Function: tiva_txpoll + * + * Description: + * The transmitter is available, check if uIP has any outgoing packets ready + * to send. This is a callback from devif_poll(). devif_poll() may be called: + * + * 1. When the preceding TX packet send is complete, + * 2. When the preceding TX packet send timesout and the interface is reset + * 3. During normal TX polling + * + * Parameters: + * dev - Reference to the NuttX driver state structure + * + * Returned Value: + * OK on success; a negated errno on failure + * + * Assumptions: + * + ****************************************************************************/ + +static int tiva_txpoll(struct net_driver_s *dev) +{ + struct tiva_driver_s *priv = (struct tiva_driver_s *)dev->d_private; + int ret = OK; + + /* If the polling resulted in data that should be sent out on the network, + * the field d_len is set to a value > 0. + */ + + nllvdbg("Poll result: d_len=%d\n", priv->ld_dev.d_len); + if (priv->ld_dev.d_len > 0) + { + /* Send the packet. tiva_transmit() will return zero if the + * packet was successfully handled. + */ + + DEBUGASSERT((tiva_ethin(priv, TIVA_MAC_TR_OFFSET) & MAC_TR_NEWTX) == 0) + arp_out(&priv->ld_dev); + ret = tiva_transmit(priv); + } + + /* If zero is returned, the polling will continue until all connections have + * been examined. + */ + + return ret; +} + +/**************************************************************************** + * Function: tiva_receive + * + * Description: + * An interrupt was received indicating the availability of a new RX packet + * + * Parameters: + * priv - Reference to the driver state structure + * + * Returned Value: + * None + * + * Assumptions: + * + ****************************************************************************/ + +static void tiva_receive(struct tiva_driver_s *priv) +{ + uint32_t regval; + uint8_t *dbuf; + int pktlen; + int bytesleft; + + /* Loop while there are incoming packets to be processed */ + + while ((tiva_ethin(priv, TIVA_MAC_NP_OFFSET) & MAC_NP_MASK) != 0) + { + /* Update statistics */ + + EMAC_STAT(priv, rx_packets); + + /* Copy the data data from the hardware to priv->ld_dev.d_buf. Set + * amount of data in priv->ld_dev.d_len + */ + + dbuf = priv->ld_dev.d_buf; + + /* The packet frame length begins in the LS 16-bits of the first + * word from the FIFO followed by the Ethernet header beginning + * in the MS 16-bits of the first word. + * + * Pick off the packet length from the first word. This packet length + * includes the len/type field (size 2) and the FCS (size 4). + */ + + regval = tiva_ethin(priv, TIVA_MAC_DATA_OFFSET); + pktlen = (int)(regval & 0x0000ffff); + nllvdbg("Receiving packet, pktlen: %d\n", pktlen); + + /* Check if the pktlen is valid. It should be large enough to hold + * an Ethernet header and small enough to fit entirely in the I/O + * buffer. Six is subtracted to acount for the 2-byte length/type + * and 4 byte FCS that are not copied into the uIP packet. + */ + + if (pktlen > (CONFIG_NET_ETH_MTU + 6) || pktlen <= (ETH_HDRLEN + 6)) + { + int wordlen; + + /* We will have to drop this packet */ + + nlldbg("Bad packet size dropped (%d)\n", pktlen); + EMAC_STAT(priv, rx_pktsize); + + /* The number of bytes and words left to read is pktlen - 4 (including, + * the final, possibly partial word) because we've already read 4 bytes. + */ + + wordlen = (pktlen - 1) >> 2; + + /* Read and discard the remaining words in the FIFO */ + + while (wordlen--) + { + (void)tiva_ethin(priv, TIVA_MAC_DATA_OFFSET); + } + + /* Check for another packet */ + + continue; + } + + /* Save the first two bytes from the first word */ + + *dbuf++ = (uint8_t)((regval >> 16) & 0xff); + *dbuf++ = (uint8_t)((regval >> 24) & 0xff); + + /* Read all of the whole, 32-bit values in the middle of the packet. + * We've already read the length (2 bytes) plus the first two bytes + * of data. + */ + + for (bytesleft = pktlen - 4; bytesleft > 7; bytesleft -= 4, dbuf += 4) + { + /* Transfer a whole word to the user buffer. Note, the user + * buffer may be un-aligned. + */ + + *(uint32_t*)dbuf = tiva_ethin(priv, TIVA_MAC_DATA_OFFSET); + } + + /* Handle the last, partial word in the FIFO (0-3 bytes) and discard + * the 4-byte FCS. + */ + + for (; bytesleft > 0; bytesleft -= 4) + { + /* Read the last word. And transfer all but the last four + * bytes of the FCS into the user buffer. + */ + + regval = tiva_ethin(priv, TIVA_MAC_DATA_OFFSET); + switch (bytesleft) + { + default: + break; + + case 7: + dbuf[2] = (regval >> 16) & 0xff; + case 6: + dbuf[1] = (regval >> 8) & 0xff; + case 5: + dbuf[0] = regval & 0xff; + break; + } + } + + /* Pass the packet length to uIP MINUS 2 bytes for the length and + * 4 bytes for the FCS. + */ + + priv->ld_dev.d_len = pktlen - 6; + tiva_dumppacket("Received packet", priv->ld_dev.d_buf, priv->ld_dev.d_len); + + /* We only accept IP packets of the configured type and ARP packets */ + +#ifdef CONFIG_NET_IPv6 + if (ETHBUF->type == HTONS(ETHTYPE_IP6)) +#else + if (ETHBUF->type == HTONS(ETHTYPE_IP)) +#endif + { + nllvdbg("IP packet received (%02x)\n", ETHBUF->type); + EMAC_STAT(priv, rx_ip); + + arp_ipin(&priv->ld_dev); + devif_input(&priv->ld_dev); + + /* If the above function invocation resulted in data that should be + * sent out on the network, the field d_len will set to a value > 0. + */ + + if (priv->ld_dev.d_len > 0) + { + arp_out(&priv->ld_dev); + tiva_transmit(priv); + } + } + else if (ETHBUF->type == htons(ETHTYPE_ARP)) + { + nllvdbg("ARP packet received (%02x)\n", ETHBUF->type); + EMAC_STAT(priv, rx_arp); + + arp_arpin(&priv->ld_dev); + + /* If the above function invocation resulted in data that should be + * sent out on the network, the field d_len will set to a value > 0. + */ + + if (priv->ld_dev.d_len > 0) + { + tiva_transmit(priv); + } + } +#ifdef CONFIG_DEBUG + else + { + nlldbg("Unsupported packet type dropped (%02x)\n", htons(ETHBUF->type)); + EMAC_STAT(priv, rx_dropped); + } +#endif + } +} + +/**************************************************************************** + * Function: tiva_txdone + * + * Description: + * An interrupt was received indicating that the last TX packet(s) is done + * + * Parameters: + * priv - Reference to the driver state structure + * + * Returned Value: + * None + * + * Assumptions: + * + ****************************************************************************/ + +static void tiva_txdone(struct tiva_driver_s *priv) +{ + /* Cancel the TX timeout */ + + wd_cancel(priv->ld_txtimeout); + + /* Verify that the Tx FIFO is not in use. The NEWTX bit initiates an + * Ethernet transmission once the packet has been placed in the TX FIFO. + * This bit is cleared once the transmission has been completed. Since + * we get here because of of TXEMP which indicates that the packet was + * transmitted and that the TX FIFO is empty, NEWTX should always be zero + * at this point. + */ + + DEBUGASSERT((tiva_ethin(priv, TIVA_MAC_TR_OFFSET) & MAC_TR_NEWTX) == 0) + + /* Then poll uIP for new XMIT data */ + + (void)devif_poll(&priv->ld_dev, tiva_txpoll); +} + +/**************************************************************************** + * Function: tiva_interrupt + * + * Description: + * Hardware interrupt handler + * + * Parameters: + * irq - Number of the IRQ that generated the interrupt + * context - Interrupt register state save info (architecture-specific) + * + * Returned Value: + * OK on success + * + * Assumptions: + * + ****************************************************************************/ + +static int tiva_interrupt(int irq, FAR void *context) +{ + register struct tiva_driver_s *priv; + uint32_t ris; + +#if TIVA_NETHCONTROLLERS > 1 +# error "A mechanism to associate and interface with an IRQ is needed" +#else + priv = &g_lm3sdev[0]; +#endif + + /* Read the raw interrupt status register */ + + ris = tiva_ethin(priv, TIVA_MAC_RIS_OFFSET); + + /* Clear all pending interrupts */ + + tiva_ethout(priv, TIVA_MAC_IACK_OFFSET, ris); + + /* Check for errors */ + +#if defined(CONFIG_DEBUG) && defined(CONFIG_DEBUG_NET) + if ((ris & MAC_RIS_TXER) != 0) + { + EMAC_STAT(priv, tx_errors); /* Number of Tx errors */ + } + + if ((ris & MAC_RIS_FOV) != 0) + { + EMAC_STAT(priv, rx_ovrerrors); /* Number of Rx FIFO overrun errors */ + } + + if ((ris & MAC_RIS_RXER) != 0) + { + EMAC_STAT(priv, rx_errors); /* Number of Rx errors */ + } +#endif + + /* Handle (unmasked) interrupts according to status bit settings */ + + ris &= tiva_ethin(priv, TIVA_MAC_IM_OFFSET); + + /* Is this an Rx interrupt (meaning that a packet has been received)? */ + + if ((ris & MAC_RIS_RXINT) != 0) + { + /* Handle the incoming packet */ + + EMAC_STAT(priv, rx_int); + tiva_receive(priv); + } + + /* Is this an Tx interrupt (meaning that the Tx FIFO is empty)? */ + + if ((ris & MAC_RIS_TXEMP) != 0) + { + /* Handle the complete of the transmission */ + + EMAC_STAT(priv, tx_int); + tiva_txdone(priv); + } + + /* Enable Ethernet interrupts (perhaps excluding the TX done interrupt if + * there are no pending transmissions). + */ + + return OK; +} + +/**************************************************************************** + * Function: tiva_txtimeout + * + * Description: + * Our TX watchdog timed out. Called from the timer interrupt handler. + * The last TX never completed. Reset the hardware and start again. + * + * Parameters: + * argc - The number of available arguments + * arg - The first argument + * + * Returned Value: + * None + * + * Assumptions: + * + ****************************************************************************/ + +static void tiva_txtimeout(int argc, uint32_t arg, ...) +{ + struct tiva_driver_s *priv = (struct tiva_driver_s *)arg; + + /* Increment statistics */ + + nlldbg("Tx timeout\n"); + EMAC_STAT(priv, tx_timeouts); + + /* Then reset the hardware */ + + DEBUGASSERT(priv->ld_bifup); + tiva_ifdown(&priv->ld_dev); + tiva_ifup(&priv->ld_dev); + + /* Then poll uIP for new XMIT data */ + + (void)devif_poll(&priv->ld_dev, tiva_txpoll); +} + +/**************************************************************************** + * Function: tiva_polltimer + * + * Description: + * Periodic timer handler. Called from the timer interrupt handler. + * + * Parameters: + * argc - The number of available arguments + * arg - The first argument + * + * Returned Value: + * None + * + * Assumptions: + * + ****************************************************************************/ + +static void tiva_polltimer(int argc, uint32_t arg, ...) +{ + struct tiva_driver_s *priv = (struct tiva_driver_s *)arg; + + /* Check if we can send another Tx packet now. The NEWTX bit initiates an + * Ethernet transmission once the packet has been placed in the TX FIFO. + * This bit is cleared once the transmission has been completed. + * + * NOTE: This can cause missing poll cycles and, hence, some timing + * inaccuracies. + */ + + if ((tiva_ethin(priv, TIVA_MAC_TR_OFFSET) & MAC_TR_NEWTX) == 0) + { + /* If so, update TCP timing states and poll uIP for new XMIT data */ + + (void)devif_timer(&priv->ld_dev, tiva_txpoll, TIVA_POLLHSEC); + + /* Setup the watchdog poll timer again */ + + (void)wd_start(priv->ld_txpoll, TIVA_WDDELAY, tiva_polltimer, 1, arg); + } +} + +/**************************************************************************** + * Function: tiva_ifup + * + * Description: + * NuttX Callback: Bring up the Ethernet interface when an IP address is + * provided + * + * Parameters: + * dev - Reference to the NuttX driver state structure + * + * Returned Value: + * None + * + * Assumptions: + * + ****************************************************************************/ + +static int tiva_ifup(struct net_driver_s *dev) +{ + struct tiva_driver_s *priv = (struct tiva_driver_s *)dev->d_private; + irqstate_t flags; + uint32_t regval; + uint32_t div; + uint16_t phyreg; + + nlldbg("Bringing up: %d.%d.%d.%d\n", + dev->d_ipaddr & 0xff, (dev->d_ipaddr >> 8) & 0xff, + (dev->d_ipaddr >> 16) & 0xff, dev->d_ipaddr >> 24 ); + + /* Enable and reset the Ethernet controller */ + + flags = irqsave(); + tiva_ethreset(priv); + + /* Set the management clock divider register for access to the PHY + * register set. The MDC clock is divided down from the system clock per: + * + * MDCCLK_FREQUENCY = SYSCLK_FREQUENCY / (2 * (div + 1)) + * div = (SYSCLK_FREQUENCY / 2 / MDCCLK_FREQUENCY) - 1 + * + * Where the maximum value for MDCCLK_FREQUENCY is 2,500,000. We will + * add 1 to assure the max TIVA_MAX_MDCCLK is not exceeded. + */ + + div = SYSCLK_FREQUENCY / 2 / TIVA_MAX_MDCCLK; + tiva_ethout(priv, TIVA_MAC_MDV_OFFSET, div); + nllvdbg("MDV: %08x\n", div); + + /* Then configure the Ethernet Controller for normal operation + * + * Setup the transmit control register (Full duplex, TX CRC Auto Generation, + * TX Padding Enabled). + */ + + regval = tiva_ethin(priv, TIVA_MAC_TCTL_OFFSET); + regval &= ~TIVA_TCTCL_CLRBITS; + regval |= TIVA_TCTCL_SETBITS; + tiva_ethout(priv, TIVA_MAC_TCTL_OFFSET, regval); + nllvdbg("TCTL: %08x\n", regval); + + /* Setup the receive control register (Disable multicast frames, disable + * promiscuous mode, disable bad CRC rejection). + */ + + regval = tiva_ethin(priv, TIVA_MAC_RCTL_OFFSET); + regval &= ~TIVA_RCTCL_CLRBITS; + regval |= TIVA_RCTCL_SETBITS; + tiva_ethout(priv, TIVA_MAC_RCTL_OFFSET, regval); + nllvdbg("RCTL: %08x\n", regval); + + /* Setup the time stamp configuration register */ + +#ifdef TIVA_ETHTS + regval = tiva_ethin(priv, TIVA_MAC_TS_OFFSET); +#ifdef CONFIG_TIVA_TIMESTAMP + regval |= MAC_TS_EN; +#else + regval &= ~(MAC_TS_EN); +#endif + tiva_ethout(priv, TIVA_MAC_TS_OFFSET, regval); + nllvdbg("TS: %08x\n", regval); +#endif + + /* Wait for the link to come up. This following is not very conservative + * of system resources -- it really should wait gracefully on a semaphore + * and the interrupt handler should post the semaphore when LINKSTATUS is + * set + */ + + nlldbg("Waiting for link\n"); + do + { + phyreg = tiva_phyread(priv, MII_MSR); + } + while ((phyreg & MII_MSR_LINKSTATUS) == 0); + nlldbg("Link established\n"); + + /* Reset the receive FIFO */ + + regval = tiva_ethin(priv, TIVA_MAC_RCTL_OFFSET); + regval |= MAC_RCTL_RSTFIFO; + tiva_ethout(priv, TIVA_MAC_RCTL_OFFSET, regval); + + /* Enable the Ethernet receiver */ + + regval = tiva_ethin(priv, TIVA_MAC_RCTL_OFFSET); + regval |= MAC_RCTL_RXEN; + tiva_ethout(priv, TIVA_MAC_RCTL_OFFSET, regval); + + /* Enable the Ethernet transmitter */ + + regval = tiva_ethin(priv, TIVA_MAC_TCTL_OFFSET); + regval |= MAC_TCTL_TXEN; + tiva_ethout(priv, TIVA_MAC_TCTL_OFFSET, regval); + + /* Reset the receive FIFO (again) */ + + regval = tiva_ethin(priv, TIVA_MAC_RCTL_OFFSET); + regval |= MAC_RCTL_RSTFIFO; + tiva_ethout(priv, TIVA_MAC_RCTL_OFFSET, regval); + + /* Enable the Ethernet interrupt */ + +#if TIVA_NETHCONTROLLERS > 1 + up_enable_irq(priv->irq); +#else + up_enable_irq(TIVA_IRQ_ETHCON); +#endif + + /* Enable the Ethernet RX packet receipt interrupt */ + + regval = tiva_ethin(priv, TIVA_MAC_IM_OFFSET); + regval |= MAC_IM_RXINTM; + tiva_ethout(priv, TIVA_MAC_IM_OFFSET, regval); + + /* Program the hardware with it's MAC address (for filtering) */ + + regval = (uint32_t)priv->ld_dev.d_mac.ether_addr_octet[3] << 24 | + (uint32_t)priv->ld_dev.d_mac.ether_addr_octet[2] << 16 | + (uint32_t)priv->ld_dev.d_mac.ether_addr_octet[1] << 8 | + (uint32_t)priv->ld_dev.d_mac.ether_addr_octet[0]; + tiva_ethout(priv, TIVA_MAC_IA0_OFFSET, regval); + + regval = (uint32_t)priv->ld_dev.d_mac.ether_addr_octet[5] << 8 | + (uint32_t)priv->ld_dev.d_mac.ether_addr_octet[4]; + tiva_ethout(priv, TIVA_MAC_IA1_OFFSET, regval); + + /* Set and activate a timer process */ + + (void)wd_start(priv->ld_txpoll, TIVA_WDDELAY, tiva_polltimer, 1, (uint32_t)priv); + + priv->ld_bifup = true; + irqrestore(flags); + return OK; +} + +/**************************************************************************** + * Function: tiva_ifdown + * + * Description: + * NuttX Callback: Stop the interface. The only way to restore normal + * behavior is to call tiva_ifup(). + * + * Parameters: + * dev - Reference to the NuttX driver state structure + * + * Returned Value: + * None + * + * Assumptions: + * + ****************************************************************************/ + +static int tiva_ifdown(struct net_driver_s *dev) +{ + struct tiva_driver_s *priv = (struct tiva_driver_s *)dev->d_private; + irqstate_t flags; + uint32_t regval; + + nlldbg("Taking down: %d.%d.%d.%d\n", + dev->d_ipaddr & 0xff, (dev->d_ipaddr >> 8) & 0xff, + (dev->d_ipaddr >> 16) & 0xff, dev->d_ipaddr >> 24 ); + + /* Cancel the TX poll timer and TX timeout timers */ + + flags = irqsave(); + wd_cancel(priv->ld_txpoll); + wd_cancel(priv->ld_txtimeout); + + /* Disable the Ethernet interrupt */ + +#if TIVA_NETHCONTROLLERS > 1 + up_disable_irq(priv->irq); +#else + up_disable_irq(TIVA_IRQ_ETHCON); +#endif + + /* Disable all Ethernet controller interrupt sources */ + + regval = tiva_ethin(priv, TIVA_MAC_IM_OFFSET); + regval &= ~MAC_IM_ALLINTS; + tiva_ethout(priv, TIVA_MAC_IM_OFFSET, regval); + + /* Reset the receive FIFO */ + + regval = tiva_ethin(priv, TIVA_MAC_RCTL_OFFSET); + regval |= MAC_RCTL_RSTFIFO; + tiva_ethout(priv, TIVA_MAC_RCTL_OFFSET, regval); + + /* Disable the Ethernet receiver */ + + regval = tiva_ethin(priv, TIVA_MAC_RCTL_OFFSET); + regval &= ~MAC_RCTL_RXEN; + tiva_ethout(priv, TIVA_MAC_RCTL_OFFSET, regval); + + /* Disable the Ethernet transmitter */ + + regval = tiva_ethin(priv, TIVA_MAC_RCTL_OFFSET); + regval &= ~MAC_TCTL_TXEN; + tiva_ethout(priv, TIVA_MAC_TCTL_OFFSET, regval); + + /* Reset the receive FIFO (again) */ + + regval = tiva_ethin(priv, TIVA_MAC_RCTL_OFFSET); + regval |= MAC_RCTL_RSTFIFO; + tiva_ethout(priv, TIVA_MAC_RCTL_OFFSET, regval); + + /* Clear any pending interrupts */ + + regval = tiva_ethin(priv, TIVA_MAC_RIS_OFFSET); + tiva_ethout(priv, TIVA_MAC_IACK_OFFSET, regval); + + /* The interface is now DOWN */ + + priv->ld_bifup = false; + irqrestore(flags); + return OK; +} + +/**************************************************************************** + * Function: tiva_txavail + * + * Description: + * Driver callback invoked when new TX data is available. This is a + * stimulus perform an out-of-cycle poll and, thereby, reduce the TX + * latency. + * + * Parameters: + * dev - Reference to the NuttX driver state structure + * + * Returned Value: + * None + * + * Assumptions: + * Called in normal user mode + * + ****************************************************************************/ + +static int tiva_txavail(struct net_driver_s *dev) +{ + struct tiva_driver_s *priv = (struct tiva_driver_s *)dev->d_private; + irqstate_t flags; + + /* Ignore the notification if the interface is not yet up or if the Tx FIFO + * hardware is not available at this time. The NEWTX bit initiates an + * Ethernet transmission once the packet has been placed in the TX FIFO. + * This bit is cleared once the transmission has been completed. When the + * transmission completes, tiva_txdone() will be called and the Tx polling + * will occur at that time. + */ + + flags = irqsave(); + if (priv->ld_bifup && (tiva_ethin(priv, TIVA_MAC_TR_OFFSET) & MAC_TR_NEWTX) == 0) + { + /* If the interface is up and we can use the Tx FIFO, then poll uIP + * for new Tx data + */ + + (void)devif_poll(&priv->ld_dev, tiva_txpoll); + } + + irqrestore(flags); + return OK; +} + +/**************************************************************************** + * Function: tiva_addmac + * + * Description: + * NuttX Callback: Add the specified MAC address to the hardware multicast + * address filtering + * + * Parameters: + * dev - Reference to the NuttX driver state structure + * mac - The MAC address to be added + * + * Returned Value: + * None + * + * Assumptions: + * + ****************************************************************************/ + +#ifdef CONFIG_NET_IGMP +static int tiva_addmac(struct net_driver_s *dev, FAR const uint8_t *mac) +{ + FAR struct tiva_driver_s *priv = (FAR struct tiva_driver_s *)dev->d_private; + + /* Add the MAC address to the hardware multicast routing table */ + +#warning "Multicast MAC support not implemented" + return OK; +} +#endif + +/**************************************************************************** + * Function: tiva_rmmac + * + * Description: + * NuttX Callback: Remove the specified MAC address from the hardware multicast + * address filtering + * + * Parameters: + * dev - Reference to the NuttX driver state structure + * mac - The MAC address to be removed + * + * Returned Value: + * None + * + * Assumptions: + * + ****************************************************************************/ + +#ifdef CONFIG_NET_IGMP +static int tiva_rmmac(struct net_driver_s *dev, FAR const uint8_t *mac) +{ + FAR struct tiva_driver_s *priv = (FAR struct tiva_driver_s *)dev->d_private; + + /* Add the MAC address to the hardware multicast routing table */ + +#warning "Multicast MAC support not implemented" + return OK; +} +#endif + +/**************************************************************************** + * Public Functions + ****************************************************************************/ + +/**************************************************************************** + * Function: tiva_ethinitialize + * + * Description: + * Initialize the Ethernet driver for one interface + * + * Parameters: + * None + * + * Returned Value: + * OK on success; Negated errno on failure. + * + * Assumptions: + * + ****************************************************************************/ + +#if TIVA_NETHCONTROLLERS > 1 +int tiva_ethinitialize(int intf) +#else +static inline int tiva_ethinitialize(int intf) +#endif +{ + struct tiva_driver_s *priv = &g_lm3sdev[intf]; + int ret; + + /* Check if the Ethernet module is present */ + + ndbg("Setting up eth%d\n", intf); + +#if TIVA_NETHCONTROLLERS > 1 +# error "This debug check only works with one interface" +#else + DEBUGASSERT((getreg32(TIVA_SYSCON_DC4) & (SYSCON_DC4_EMAC0|SYSCON_DC4_EPHY0)) == (SYSCON_DC4_EMAC0|SYSCON_DC4_EPHY0)); +#endif + DEBUGASSERT((unsigned)intf < TIVA_NETHCONTROLLERS); + + /* Initialize the driver structure */ + + memset(priv, 0, sizeof(struct tiva_driver_s)); + priv->ld_dev.d_ifup = tiva_ifup; /* I/F down callback */ + priv->ld_dev.d_ifdown = tiva_ifdown; /* I/F up (new IP address) callback */ + priv->ld_dev.d_txavail = tiva_txavail; /* New TX data callback */ +#ifdef CONFIG_NET_IGMP + priv->ld_dev.d_addmac = tiva_addmac; /* Add multicast MAC address */ + priv->ld_dev.d_rmmac = tiva_rmmac; /* Remove multicast MAC address */ +#endif + priv->ld_dev.d_private = (void*)priv; /* Used to recover private state from dev */ + + /* Create a watchdog for timing polling for and timing of transmisstions */ + +#if TIVA_NETHCONTROLLERS > 1 +# error "A mechanism to associate base address an IRQ with an interface is needed" + priv->ld_base = ??; /* Ethernet controller base address */ + priv->ld_irq = ??; /* Ethernet controller IRQ number */ +#endif + priv->ld_txpoll = wd_create(); /* Create periodic poll timer */ + priv->ld_txtimeout = wd_create(); /* Create TX timeout timer */ + + /* If the board can provide us with a MAC address, get the address + * from the board now. The MAC will not be applied until tiva_ifup() + * is caleld (and the MAC can be overwritten with a netdev ioctl call). + */ + +#ifdef CONFIG_TIVA_BOARDMAC + tiva_ethernetmac(&priv->ld_dev.d_mac); +#endif + + /* Perform minimal, one-time initialization -- just reset the controller and + * leave it disabled. The Ethernet controller will be reset and properly + * re-initialized each time tiva_ifup() is called. + */ + + tiva_ethreset(priv); + tiva_ifdown(&priv->ld_dev); + + /* Attach the IRQ to the driver */ + +#if TIVA_NETHCONTROLLERS > 1 + ret = irq_attach(priv->irq, tiva_interrupt); +#else + ret = irq_attach(TIVA_IRQ_ETHCON, tiva_interrupt); +#endif + if (ret != 0) + { + /* We could not attach the ISR to the IRQ */ + + return -EAGAIN; + } + + /* Register the device with the OS so that socket IOCTLs can be performed */ + + (void)netdev_register(&priv->ld_dev, NET_LL_ETHERNET); + return OK; +} + + +/************************************************************************************ + * Name: up_netinitialize + * + * Description: + * Initialize the first network interface. If there are more than one interface + * in the chip, then board-specific logic will have to provide this function to + * determine which, if any, Ethernet controllers should be initialized. + * + ************************************************************************************/ + +#if TIVA_NETHCONTROLLERS == 1 +void up_netinitialize(void) +{ + (void)tiva_ethinitialize(0); +} +#endif + +#endif /* CONFIG_NET && CONFIG_TIVA_ETHERNET */ |